Just uploaded to CPAN, this has been a pet project of mine for a while (see perlmonks here ). It is currently only a alpha release, and you know what that means.
Converting UNIX shell scripts creates problems and my script (with associated modules) do not answer them all. The syntax of shell programs is very loose, almost anarchic. This is not a criticism of shells - they have a dual role to play in providing a usable command-line interface, as well as acting as a programming language. I target the POSIX shell, but there are many variations, not least those created by Sun. The Korn shell has moved on considerably from ksh88, and the syntax of the latest versions is complex. I make no attempt at the more esoteric parts of ksh93 syntax, expansions and extended globbing. There are annoying inconsistencies, fore example, what is the difference between:
[ $x -a $b ]
and
[[ $x -a $b ]]
I shall leave the reader to ponder (hint: -a has a different meaning to the Bourne-shell [ and the Korn shell [[).
As you might gather, this is not a simple syntax conversion. Take $? for example: that can be applied to any shell command you like and at any point in the program. Not many languages work like that, certainly Perl does not. Fortunately not many people use $? in shell scripts for anything other than testing the result of a child process - not unlike its use in Perl (there is a difference in the detail).
Right now App::sh2p attempts to remove some of the leg-work from conversion, and I hope t is useful.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment